Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add aaronartistzhang-afk/DailyWork --skill agent-fleet-guardgit clone --depth 1 https://github.com/aaronartistzhang-afk/DailyWorkWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/aaronartistzhang-afk/dailywork/agent-fleet-guard)<a href="https://agentmods.dev/skills/aaronartistzhang-afk/dailywork/agent-fleet-guard"><img src="https://agentmods.dev/badge/skills/aaronartistzhang-afk/dailywork/agent-fleet-guard/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/aaronartistzhang-afk/dailywork/agent-fleet-guard"><img src="https://agentmods.dev/badge/skills/aaronartistzhang-afk/dailywork/agent-fleet-guard.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00172 | $0.01091 |
| Opus 5.5 | $0.00069 | $0.00436 |
| Sonnet 5.5 | $0.00034 | $0.00218 |
| Haiku 4.5 | $0.00017 | $0.00109 |
Grade A, and why
agent-fleet-guard scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 59 lines — stays where its author put it; the contents beside it link to each section on GitHub.
agent-fleet-guard — guardrails for large agent orchestrations
Dispatching one agent is routine. Dispatching twenty is a spend decision — and a fleet that half-dies silently poisons everything downstream. Six guardrails, all paid for with real incidents.
Threshold: fan-out ≥ 20 subagents, or estimated ≥ 1M subagent tokens. Below that, just dispatch.
① Quote gate — quote first, wait for approval
Before dispatch, run the estimator (same directory):
./fleet_estimate.sh --agents <N> --per <rough-tokens-per-agent> --json
Exit 10 = over threshold (agents ≥ 20 or corrected estimate ≥ 1M) → you must quote and wait:
"Planning N agents, rough estimate X tokens, ×3 corrected ≈ Y tokens (both are floors, not ceilings) — the output is Z. Approve?"
Give both numbers — the rough estimate and the ×3 correction. Why ×3: a real fleet quoted at ~0.9M tokens burned ~2.7M — roughly 3×, and that was a lower bound. Per-agent guesses run low; the estimator's default --factor 3 bakes this in.
② Don't launch on an unstable connection
If this session has recently seen Connection closed / connection-lost errors: treat that as the signal, no probing needed — crashed agents burn their tokens for nothing, and fail-closed pipelines then demand a full rerun. The classic failure is seeing scattered disconnects and not counting them as a signal.
No existing signal but the fan-out is over threshold? Send 3 one-token probe agents first; any probe timing out >10s or erroring → unstable. Unstable → don't force it: split into smaller batches or wait.
③ In-flight stop-loss
While the fleet runs, track cumulative spend (orchestration frameworks expose a budget/spent counter; with散派 agents, sum their usage by hand). The moment spend crosses the approved number Y — stop and re-quote, without waiting to be asked.
Anchor on the approved number, not the rough estimate — you quoted ×3, so anchoring on the rough estimate false-alarms at ⅓ of every normal run.
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 59 lines · 172 tokens per session scan A 3fa5ca0b22c0
agent-fleet-guard is a skill published in the GitHub repository aaronartistzhang-afk/DailyWork (1 stars, last pushed 1mo ago), licensed MIT. It adds 172 tokens to every session and 1,091 once invoked, about $0.0007 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
uig
Build a User Input Grid (UIG) for an AI product or feature, evaluate existing eval datasets against it, and propose new inputs to fill coverage gaps. Use this skill whenever someone asks to design an eval framework, audit a test set, build a "user input grid", "synthetic query matrix" or "SQM", improve dataset…
eval-llm-judge
Write LLM-as-judge evaluator prompts for AI products. Use this skill whenever you need to evaluate semantic properties of AI outputs that cannot be checked programmatically — output quality, tone, factual accuracy, edge case handling, reference alignment. Also use when auditing existing judge prompts for bias…
eval-code
Write deterministic, code-based evaluator functions for an AI product. Use this skill whenever you need to write evaluators that check structural properties of AI outputs (format, count, presence, schema compliance) without calling another LLM. Also use when auditing existing code-based evaluators for brittleness…
llm-align
Analyze alignment between LLM judge scores and human labels in an eval dataset. Use this skill whenever someone wants to evaluate how well an LLM judge agrees with human reviewers, calculate TPR/TNR, investigate disagreements, or improve a scorer prompt. Trigger on phrases like: "calculate TPR/TNR", "judge alignment"…
ticket-to-eval
Strip PII from a customer support ticket or eval trace and convert it into eval dataset rows. Produces two outputs: a regression dataset row (close to the original input, tagged with failuremode) and a generalized dataset row (abstracted for broader coverage). Works with local CSV files or any eval platform. Use this…
trace-codes
Turn free-form notes from reading agent traces into trace codes: named, yes/no failure or quality patterns with definitions, examples and counts. Use this skill whenever someone has annotated traces or review notes and asks to "cluster my notes", "create trace codes", "find the failure modes", "do error analysis on…